EP-4741179-A1 - BATTERY LIFE PREDICTION DEVICE AND BATTERY LIFE PREDICTION METHOD

EP4741179A1EP 4741179 A1EP4741179 A1EP 4741179A1EP-4741179-A1

Abstract

A battery life prediction apparatus is for predicting the life of a battery of a detection apparatus that detects information on the state of a tire. The battery life prediction apparatus includes an acquisition interface (131) that acquires input data including tire temperature, tire pressure, a value related to voltage of a battery, and a battery usage time, and a predictor (133) that predicts, as the life of the battery, a value related to the voltage after a predetermined future period using the input data and a prediction model. The prediction model is generated by machine learning using training data that takes the tire temperature, the tire pressure, the value related to the voltage of the battery, and the usage time of the battery in performance data as explanatory variables, and a value related to the voltage of the battery after the predetermined period as an objective variable.

Inventors

GOTO, Takato
KURAMOTO, Toshiki
TAKASAWA, Yuya

Assignees

BRIDGESTONE CORPORATION

Dates

Publication Date: 20260513
Application Date: 20240729

Claims (9)

A battery life prediction apparatus for predicting a life of a battery of a detection apparatus that detects information on a state of a tire, the battery life prediction apparatus comprising: an acquisition interface configured to acquire input data including a temperature of the tire, a pressure of the tire, a value related to a voltage of the battery, and a usage time of the battery; and a predictor configured to predict, as the life of the battery, a value related to the voltage after a predetermined period in the future, using the input data and a prediction model, wherein the prediction model is generated by machine learning using training data that takes the temperature of the tire, the pressure of the tire, the value related to the voltage of the battery, and the usage time of the battery in performance data as explanatory variables, and a value related to the voltage of the battery after the predetermined period as an objective variable.
The battery life prediction apparatus according to claim 1, wherein the input data is acquired from the detection apparatus by fixed-point observation by a reading apparatus installed at a predetermined location, and the acquisition interface is configured to acquire the input data via the reading apparatus.
The battery life prediction apparatus according to claim 1 or 2, wherein the value related to the voltage is a voltage value that is segmented and grouped, and at least one group corresponds to a state of the battery when the battery is to be replaced.
The battery life prediction apparatus according to claim 3, wherein the usage time of the battery is an elapsed time from when the value related to the voltage first fluctuates.
The battery life prediction apparatus according to any one of claims 1 to 4, wherein the performance data is stored in a storage apparatus on a network as viewed from the battery life prediction apparatus.
The battery life prediction apparatus according to claim 5, wherein the prediction model and the predicted value related to the voltage are stored in the storage apparatus.
The battery life prediction apparatus according to any one of claims 1 to 6, wherein the predicted value related to the voltage is provided to a user via a network.
The battery life prediction apparatus according to any one of claims 1 to 7, wherein the training data uses a basic statistical quantity with regard to the temperature of the tire and the pressure of the tire in the performance data as the explanatory variables.
A battery life prediction method for predicting a life of a battery of a detection apparatus that detects information on a state of a tire, the battery life prediction method comprising: acquiring input data including a temperature of the tire, a pressure of the tire, a value related to a voltage of the battery, and a usage time of the battery; and predicting, as the life of the battery, a value related to the voltage after a predetermined period in the future, using the input data and a prediction model, wherein the prediction model is generated by machine learning using training data that takes the temperature of the tire, the pressure of the tire, the value related to the voltage of the battery, and the usage time of the battery in performance data as explanatory variables, and a value related to the voltage of the battery after the predetermined period as an objective variable.

Description

TECHNICAL FIELD The present disclosure relates to a battery life prediction apparatus and a battery life prediction method. BACKGROUND Conventionally, various techniques for predicting battery life have been proposed. For example, PTL 1 discloses a technique for obtaining the remaining life of a storage battery for an electric vehicle. In the technique of PTL 1, regression analysis is performed using a plurality of sets of accumulated travel counts and corresponding voltage values to obtain a regression equation. From the regression equation, the travel count corresponding to the life voltage value at a predetermined discharge amount is obtained, and the current travel count is subtracted to estimate the remaining life travel count. CITATION LIST Patent Literature PTL 1: JP H06-163084 A SUMMARY (Technical Problem) Here, there is a demand for a technique for predicting the life of a battery used in devices mounted on vehicles other than electric vehicles. For example, sensors and transmission units used in tire pressure monitoring systems (TPMS: Tire Pressure Monitoring System) are often powered by batteries. In order to prevent the data transmission of the TPMS, which is necessary for managing the state of the tire, from stopping, predicting the battery life and replacing it at an appropriate timing is important from the viewpoint of safe driving of the vehicle. The technique of PTL 1 estimates how many times remain until the capacity that can be fully charged becomes the usable line as a storage battery for an electric vehicle, because the capacity that can be fully charged decreases as the storage battery deteriorates due to repeated charging and discharging. That is, the technique of PTL 1 predicts changes in the fully charged capacity due to repeated charging and discharging, and since the prediction target is different, it cannot be applied to battery life prediction (estimation of remaining amount) for TPMS. In view of such circumstances, an object of the present disclosure is to provide a battery life prediction apparatus and a battery life prediction method capable of highly accurately predicting the battery life of a detection apparatus that detects information on the state of a tire. (Solution to Problem) (1) A battery life prediction apparatus according to one embodiment of the present disclosure is a battery life prediction apparatus that predicts the life of a battery of a detection apparatus that detects information on the state of a tire, comprising: an acquisition interface configured to acquire input data including a temperature of the tire, a pressure of the tire, a value related to a voltage of the battery, and a usage time of the battery; anda predictor configured to predict, as the life of the battery, a value related to the voltage after a predetermined period in the future, using the input data and a prediction model,wherein the prediction model is generated by machine learning using training data that takes the temperature of the tire, the pressure of the tire, the value related to the voltage of the battery, and the usage time of the battery in performance data as explanatory variables, and a value related to the voltage of the battery after the predetermined period as an objective variable.(2) In one embodiment of the present disclosure, in (1), the input data is acquired from the detection apparatus by fixed-point observation by a reading apparatus installed at a predetermined location, andthe acquisition interface is configured to acquire the input data via the reading apparatus.(3) In one embodiment of the present disclosure, in (1) or (2), the value related to the voltage is a voltage value that is segmented and grouped, and at least one group corresponds to a state of the battery when the battery is to be replaced.(4) In one embodiment of the present disclosure, in (3), the usage time of the battery is an elapsed time from when the value related to the voltage first fluctuates.(5) In one embodiment of the present disclosure, in any one of (1) to (4), the performance data is stored in a storage apparatus on a network as viewed from the battery life prediction apparatus.(6) In one embodiment of the present disclosure, in (5), the prediction model and the predicted value related to the voltage are stored in the storage apparatus.(7) In one embodiment of the present disclosure, in any one of (1) to (6), the predicted value related to the voltage is provided to a user via a network.(8) In one embodiment of the present disclosure, in any one of (1) to (7), the training data uses a basic statistical quantity with regard to the temperature of the tire and the pressure of the tire in the performance data as the explanatory variables.(9) A battery life prediction method according to one embodiment of the present disclosure is a battery life prediction method for predicting the life of a battery of a detection apparatus that detects information on the state of a tire, comprising: acquiri